Summary of Multi-agents Based on Large Language Models For Knowledge-based Visual Question Answering, by Zhongjian Hu et al.
Multi-Agents Based on Large Language Models for Knowledge-based Visual Question Answering
by Zhongjian Hu, Peng Yang, Bing Li, Zhenqi Wang
First submitted to arxiv on: 24 Dec 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed multi-agent voting framework utilizes Large Language Models (LLMs) to simulate different staff levels in a team, addressing the limitations of existing VQA methods. The framework comprises three LLM-based agents that utilize external tools autonomously and collaborate to provide better answers. Each agent provides an answer based on its level, which is then voted upon to obtain the final result. This approach outperforms baselines by 2.2 and 1.0 on OK-VQA and A-OKVQA datasets respectively. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine a team of smart computers that work together to give better answers to tricky questions. Just like how we humans use tools and ask for help when we need it, these computer agents use different levels of expertise and teamwork to come up with the best answer. This new way of working helps computers answer questions more accurately than before. |
